an
g
, M
ies, K
; acce
19 S
A single-pulse TMS to the human motor cortex (M1) influences reaction time (RT). We may summarize from previous studies where
that TMS above motor threshold (MT) delays RT, whereas TMS below
1. Introduction
completely draw this conclusion from various studies where
different groups of subjects participated in various types of
RT tasks, i.e. simple RT, choice RT and GO/NOGO task.
TMS effects on M1, we may expect that single-pulse TMS
above MT delays RT, whereas TMS below MT shortens RT
Neuroscience Research 50 (2and that the TMS effects on RT depend on the timings of
TMS during RT period, even though TMS at an identical
intensity is delivered.
* Corresponding author. Tel.: +81 75 753 6887; fax: +81 75 753 7907.
E-mail address: eiichi.naito@neuro.mbox.media.kyoto-u.ac.jp
(E. Naito).
0168-0102/$ – see front matter # 2004 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.
doi:10.1016/j.neures.2004.08.002A single-pulse TMS to the human motor cortex (M1)
affects its neuronal processes. In the previous studies, where
M1 was TMS-stimulated during various types of reaction
time (RT) tasks, inconsistent results have been reported. One
is that TMS shortens RT (Pascual-Leone et al., 1992a,
1992b, 1994; Sawaki et al., 1999), and the other is that it
prolongs RT (Day et al., 1989; Romaiguere et al., 1997;
Ziemann et al., 1997; Schluter et al., 1998, 1999; Burle et al.,
2002). One may speculate that this inconsistency is most
probably due to the differences between intensities or
timings of TMS used in these studies. However, one may not
Because, first, definitions of motor threshold (MT)
methodologically varies across studies, second, a choice
RT task contains response selection process that does not
exist in a simple RT task, and if TMS is delivered during this
selection process, the effect of TMS should not be
considered as the same one when TMS is delivered during
simple RT period.
Thus, we tested these TMS effects in a single study where
both higher and lower intensities at various timings during
simple RT period are applied in an identical group of
subjects. If previous findings are general conclusions ofMT shortens RT and that these RT changes depends on TMS timings during RT period. However, these effects have never been systematically
investigated in a single study where an identical group of subjects participated. The purpose of this study is to test previous TMS effects in a
study of simple RT task. Seven subjects isometrically abducted their right index fingers as quickly as possible when a visual stimulus
appeared. A single-pulse TMS was randomly delivered over the left M1 at various timings during RT period in a single trial (at 0, 40, 60, 80 or
100 ms after the visual stimulus). Motor-evoked potential (MEP) and EMG activity for response were recorded from the right finger muscles.
Only the TMS above MT delivered at 80 or 100 ms, which increased MEP amplitude, significantly delayed RT and increased the size of
response EMG activities that may reflect contents of central motor commands. The TMS below MT at these timings, which occasionally
evoked MEP, exclusively shortened RT despite the fact that the response EMG size was unchanged. A single-pulse TMS has different effects
on the ongoing neuronal processes in M1 during the pre-movement period: TMS above MT may temporally retard the processes and also
affect contents of central motor commands, whereas TMS below MT may simply facilitate its processes without affecting motor commands.
# 2004 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.
Keywords: Reaction time; Transcranial magnetic stimulation; Motor-evoked potential; Motor cortex; Corticospinal excitability; Humandifferent groups of subjects participated in various types of RT tasksTwo different effects of transcr
human motor cortex durin
Toshihiro Hashimoto, Daisuke Inaba
Graduate School of Human and Environmental Stud
Received 9 June 2004
Available online
Abstractial magnetic stimulation to the
the pre-movement period
ichikazu Matsumura, Eiichi Naito*
yoto University, Sakyo-ku, Kyoto 606-8501, Japan
pted 10 August 2004
eptember 2004
www.elsevier.com/locate/neures
004) 427–436

circle, 5.0 mm in diameter). The duration of the visual sti-
encemulus was 50 ms. When the LED lit up green, the subjects
were instructed to isometrically abduct their right index
finger as quickly as possible. And they were also required to
generate the same finger movement in all trials of all con-
ditions. The inter-stimulus interval (ISI) was varied randomly
from 4 to 16 s to avoid an anticipation effect for the timing of
next signal. Between trials the muscles remained relaxed. In
order to minimize the variability of RTs across trials and to
make the subjects produce the same type of the finger move-
ment across trials, the subjects repeatedly practiced a simple
RT (SRT) task (96 trials) for 5 min before the experiments.
2.3. Transcranial magnetic stimulation
A single-pulse TMS was delivered by using a mono-We test this by conducting a simple visual RT task where
seven right-handed healthy male subjects isometrically
abduct their right index fingers as quickly as possible when
a visual stimulus appears. A single-pulse TMS above or
below MT is randomly delivered over the left M1 at various
timings during RT period in a single trial (at 0, 40, 60, 80 or
100 ms after the visual stimulus presentation). We measure
RT and also record motor-evoked potential (MEP) and
response muscle activities to the visual stimulus from the
right ‘first dorsal interosseous’ (FDI) muscles. We analyze
TMS effect to ongoing neuronal processes in M1 by
evaluating changes of RT and of response muscle activities
and by evaluating the amplitude and the latency of MEP from
the target muscles (Naito et al., 2002).
2. Experimental procedures
2.1. Subjects and experimental conditions
Seven healthy volunteers (all men; mean age 24.1, range
21–26 years old) participated in this experiment. All subjects
were right-handed according to the Edinburgh scale (Old-
field, 1971). None of them had prior history of neurological
disease or brain injury. The subjects had given their written
consent, which was approved by the local Ethical Committee
of Graduate School of Human and Environmental studies of
Kyoto University. TMS experiments were carried out
following the principles and guidelines of the Declaration
of Helsinki (1975). Experiments were carried out in a dimly
illuminated, shielded room. Subjects sat on a comfortable
chair, and totally relaxed. Right and left forearms were
pronated. Both forearms and hands were placed on a
horizontal platform in front of them.
2.2. Visual stimulation
A response signal was presented by a light-emitting diode
(LED) that was 50 cm in front of the subjects at eye level (a
T. Hashimoto et al. / Neurosci428phasic stimulator (Nihon Kohden SMN1200, maximumintensity was 0.67 T) through a figure-eight coil (each
external diameter 10.0 cm). The coil was held tangentially
on the scalp at the optimal ‘hot spot’ where TMS evoked
MEPs from the right FDI muscle.
2.4. Identification of TMS stimulation site
After the RT training, we identified the optimal
stimulation site on the scalp over the hand section of the
left primary motor cortex. The subjects were required to be
totally relaxed with no muscular contraction and to fixate
their eyes on the LED (Hess et al., 1987; Thompson et al.,
1991). We used 80% intensity of our TMS machine, which
produced MEPs (Naito et al., 2002). TMS was applied at a
site 5 cm lateral to the vertex (Cz, 10/20 Electrode System)
in the left hemisphere (Naito and Matsumura, 1994a, 1994b,
1996; Naito et al., 2002). Starting from this location we kept
finding a ‘hot spot’, where TMS evokes MEP in the right
‘first dorsal interosseous’ (FDI: agonistic muscles for the
response), by moving a figure-eight TMS coil anteriorly,
posteriorly, laterally or medially in a step of 1 cm. Once we
identified a ‘hot spot’, we again stimulated this at about four
times in order to confirm whether it was an optimal ‘hot
spot’ where we could evoke maximum MEPs. After this
procedure, we completely fixed the coil on the location
throughout our experiment. Orientation of the coil was fixed
in order to induce the current from posterior to anterior
(Werhahn et al., 1994).
2.5. Definition of intensity of TMS
After we fixed the coil position, we determined the TMS
intensity. Definitions of supra-threshold (above MT) and
sub-threshold (below MT) are inconsistent across previous
studies (Pascual-Leone et al., 1992a, 1992b, 1994; Day et
al., 1989; Romaiguere et al., 1997; Schluter et al., 1998,
1999; Ziemann et al., 1997; Sawaki et al., 1999; Burle et al.,
2002). In the present study, we adopted the method
introduced by Mills and Nithi (1997). We defined an ‘upper
threshold’ (UT) as the minimum intensity at which all of the
10 single-TMS stimuli produce 10 corresponding MEPs and
a ‘lower threshold’ (LT) as the maximum intensity at which
all 10 stimuli never produce any MEPs. These definitions
can provide absolutely higher or lower threshold than the
motor threshold (Rossini et al., 1994), which is usually
determined as the TMS intensity where the TMS evokes
MEP at 50% probability in 10–20 trials. We adopted this
method because in many previous studies UT or LT was
determined by simply increasing or decreasing TMS
intensity from the MT without further evaluating probability
of MEP once the UT or the LT was delivered. By applying
the present method, one may clearly determine UT or LT by
providing precise probability of MEP.
First, we defined LT. We started with 80% of the
maximum intensity (0.67 T) of our TMS machine. A single-
Research 50 (2004) 427–436pulse TMS was provided to the left M1 in one trial and we